Abstract
Increasing prevalence of dementia has posed several challenges for care-givers. Patients suffering from dementia often display wandering behavior due to boredom or memory loss. It is considered to be one of the challenging conditions to manage and understand. Traits of dementia patients can compromise their safety causing serious injuries. This paper presents investigation into the design and evaluation of wandering scenarios with patients suffering from dementia using an S-band sensing technique. This frequency band is the wireless channel commonly used to monitor and characterize different scenarios including random, lapping, and pacing movements in an indoor environment. Wandering patterns are characterized depending on the received amplitude and phase information of that measures the disturbance caused in the ideal radio signal. A secondary analysis using support vector machine is used to classify the three patterns. The results show that the proposed technique carries high classification accuracy up to 90% and has good potential for healthcare applications.
Original language | English |
---|---|
Pages (from-to) | 1863 - 1870 |
Number of pages | 8 |
Journal | IEEE Journal of Biomedical and Health Informatics |
Volume | 22 |
Issue number | 6 |
DOIs | |
Publication status | Published - 27 Dec 2017 |
Externally published | Yes |
Keywords
- S-Band sensing
- wandering patterns
- dementia
Fingerprint Dive into the research topics of 'Wandering pattern sensing at S-band'. Together they form a unique fingerprint.
Profiles
-
Syed Aziz Shah
- Faculty Research Centre for Intelligent Healthcare - Associate Professor (Research)
Person: Teaching and Research